Professor Ariel Shamir

Biography

Associate professor at the Efi Arazi school of Computer Science at the Interdisciplinary
Center in Israel.


He received his Ph.D. in computer science from the Hebrew University in Jerusalem in 2000. He has spent two years at the computational visualization center at the University of Texas in Austin, and in 2007 held the position of a visiting scientist at Mitsubishi Electric Research Labs in Cambridge MA. During the years he has been involved in consulting for a number of places such as Lawrence Livermore national laboratory, Walt Disney Research, and a number of high-tech companies in Israel. He also co-authored several patents. His research interests include geometric modeling, computer graphics, visualization, and machine learning. He is a member of the ACM SIGGRAPH, IEEE Computer, and Eurographics associations.

---

Program : A volumetric view of shapes

Shape analysis relies heavily on extracting features from the underlying geometric model. Many times the feature are affected by the model representation. For instance, curvature and planarity are useful surface properties that lend themselves well to the boundary representation of 3D models by meshes and surfaces. Still, many of the objects represented by a boundary surface are volumetric in nature, and their analysis should utilize this understanding.


In this talk I will present the shape diameter function (SDF) which is a scalar field that approximates the local shape diameter at a point but is defined on the boundary surface. This function provides a link between volumetric properties of the shape and its representation as a boundary surface. I will demonstrate the usefulness of SDF in various applications such as shape retrieval, shape segmentation, skeleton extraction, and shape classification.